Talent Systems — Science Team
Platform Overview

Platform Architecture

How the platform components fit together — what matters for science configuration.

What You Need to Know

ComponentWhat It DoesWhat You Control
AI InterviewerRetell AI + GPT-4.1 conducts real-time interviewsInterviewer rubric (tone, probing, rules, question generation)
AI ScorerClaude analyzes transcripts post-interviewScorer rubric (methodology, thresholds, anchors)
Competency FrameworkDatabase of competencies with 5-level anchorsAnchor definitions, categories, platform competencies
Integrity MonitorBrowser-based face detection + behavior trackingDeduction weights, detection thresholds
AI HelpersClaude generates JDs, competencies, questionsSystem/task prompts for generation quality

Three User Surfaces

  1. Admin Portal (/admin) — Where you configure rubrics, competencies, and all science settings
  2. Employer Portal (/dashboard) — What employers see. Postings, candidates, scorecards, team.
  3. Candidate Portal — Token-gated interview experience. Privacy → resume → interview → complete.

Key Design Decisions

  • No audio/video stored — ever. Camera is for live face detection only.
  • Scorer never sees candidate identity — name scrubbed from transcript before scoring.
  • All science configuration in databaserubrics table stores JSON configs, editable via admin UI. Nothing is hardcoded.
  • Soft-delete everywhere — EEOC, NYC 144, GDPR compliance.
  • Rubric versioning — every save creates an immutable version snapshot. Versions are stamped on transcripts and scorecards for traceability.

On this page